London: Researchers in Finland say the answers are on the Internet.
They’re adding their voices to the idea that Google data can help improve economic analysis and forecasts in areas such as the labour market.
By tracking searches for terms such as “unemployment benefits” or “career service” across the European Union (EU), the Research Institute of the Finnish Economy in Helsinki (known as ETLA) says it’s able to predict unemployment as much as three months in advance. While central banks and statistics agencies have already expressed interest—and some already use it to get additional insights—ETLA says it can now estimate real-time unemployment rates using web searches.
The newly unveiled research could have predicted joblessness in Portugal, where ETLA’s Google Index began to spike in 2011 before the unemployment rate did. That was also the case in the Netherlands, Finland and France, meaning the tool could be of interest to policy makers.
“Online activity is correlated with the real world, and what is even more important is that we can possibly use that correlation to predict the economy,” Joonas Tuhkuri, the ETLA researcher who led the project, told a conference at the European Parliament in Brussels. “There are over 100 billion Google searches every month, and each search is someone expressing an interest in or a demand for something.”
The group’s index updates in real time, providing a useful comparison to official statistics. ETLA’s nowcast as of Tuesday morning showed that unemployment is currently 4.92% in the UK, 20.52% in Spain, 10.97% in Finland and 10.16% in France. According to official estimates for the first quarter, unemployment stood at 5.1% in Britain, 21% in Spain and 10.2% in France. It was 9.8% in Finland in April.
ETLA researchers plan to extend the method to the housing market. They are also looking at Google searches in Syria as an indication of how many refugees might arrive in Europe.
The Federal Reserve, the Bank of England and central banks in Italy, Israel, Turkey and Chile saw the potential of using Internet search data for their economic analysis several years ago. The European Commission is examining the issue, and Britain’s Office for National Statistics is planning to use new methods including web scraping to better measure the economy.
But it’s not a simple case of out with the old and in with the new.
“Big data has to be on the agenda of official statistics,” said Silke Stapel-Weber, director of national accounts, prices and key indicators at Eurostat, the EU statistics agency. “The real success will come if you combine all of those things, because we are deeply convinced that you always will need official statistics as a benchmark for many other data sources.”